Antigenic drift is a change in structure of a virus, such as the internal proteins, glycoproteins, glycolipids, etc., due to fundamental change in the genomic content of the virus particle. The genome may be comprised of either RNA molecules, e.g., herpes virus type 1 or type 2, etc. Variable host cell range, diseases morbidity, and the like, are the inevitable sequence of fundamental structural changes in the compositions of given groups of viral particles.
Human immunodeficiency virus (HIV), the etiologic agent of acquired immunodeficiency syndrome (AIDS), selectively infects and destroys the host's helper subset of T-lymphocytes thereby giving rise to severe opportunistic infections and certain neoplasias (1). Monocytes have also been identified as a major target for HIV infection and may be responsible for the frequent acute encephalopathy associated with AIDS (2). A comparison of various HIV isolates has revealed a high degree of genomic diversity. Whereas isolates such as HTLV-III.sub.8 and LAV differ by less than 2%, the Haitian isolate HTLV-III.sub.RF shows much greater divergence (3). The most variable region of the HIV genome is the envelope (env) gene and variability in this region is a mechanism for escaping the host's immune response (4). However, this variation and variation that occurs throughout the remainder of the genome also reflects a potential for different efficacies of antiviral drugs against the many variants of HIV. Therefore, to be usefully predictive, anti-HIV drug testing must include diverse HIV isolates. In addition, potentially useful anti-HIV drugs can be screened for activity on monocytes as well as helper T lymphocytes since both lineages serve as major targets for human viruses in general and since both play a role in the pathogenesis of subacute/chronic infection in vivo. HIV-2 (also called LAV-2) has approximately only 60% homology with (or similarity to) HIV-1 has diverged from a common ancestor at an indeterminate past time.
Double-stranded RNAs (dsRNAs) are potent biological response modifiers having pleiotropic activities which include induction of the interferons (IFNs), activation of certain IFN-induced enzymes such as 2-5A synthetase, augmentation of natural killer and monocyte activities, and B and T cell mitogenic activities. Indeed, I have shown synthetic mismatched dsRNA of the form r(I).sub.n .multidot.r(C.sub.12 -U).sub.n (Ampligen, a trademark of HEM Research, Inc., Rockville, Md., U.S.A.) to have anti-HIV activity in vitro; see my published European patent application No. 0,213,921. Furthermore, Ampligen and azidothymidine (AZT) acted synergistically against HIV in similar infection assays; see my copending application Ser. No. 07/028,823. As there described, in a clinical trial, I demonstrated that Ampligen improved the clinical status of ten patients with AIDS-related complex (ARC), lymphadenopathy syndrome (LAS) or AIDS while producing only rare, mild side effects.
In a continuing effort to characterize the full potential efficacy of dsRNAs and related mismatched dsRNAs in the treatment of HIV infection, I have extended by previous investigations to include the highly divergent HIV isolate HTLV-III.sub.RF (5) and three additional target cell lines, including the T-lymphoblastoid cell line H9 (6), the MTLV-I-transformed T cell line MT-2 (7), and the monocyte/macrophage cell line U937 (8). From these studies, I have concluded that dsRNA under certain circumstances may prevent viral escape of HIV as a prototype virus. Perhaps more importantly, I have determined that dsRNA can continue to be effective in a milieu consisting of different genomic variants of animal viruses as well as in the presence of factors which promote viral escape, such as antibodies which actually enhance viral replication and, therefore, the potential for viral escape.
Retroviruses in general, and HIV in particular, give rise to antibody-dependent viral enhancing (ADE) factors produced by the host. Antibody-dependent enhancement of HIV infection considerably worsens the infection and accelerates the rate of patient decline. This ADE prevents the host's immune system from raising antibodies effective against the viral invader and must be adequately controlled for effective therapy. In that the presence of ADE, may prevent animals from forming adequate or effective antibodies against pathogenic agents, my invention is of critical value since by protecting the animal against the effects of ADE it allows for adequate levels of neutralizing antibody to be produced.
Like other retroviruses, HIV-1 is characterized by high antigenic variation and genomic heterogeneity. Antigenic drift and the resulting highly polymorphic variants of the virus enable it to escape the host's immune defense mechanisms and may aggravate and delay the development of an effective hetrovaccine.
The production of enhancing factors to HIV in immunized animals raises serious concern about the types of antibodies that could be elicited by vaccine approaches in man. Preliminary observations indicate that some isolates are more susceptible to enhancement than others. Such viruses could prove valuable for identifying the epitope(s) responsible for antibody-dependent enhancement that should be removed from vaccine preparations.
At least two components of human serum enhance human immunodeficiency virus type 1 (HIV-1) infection and mask HIV-1 neutralizing antibody activity; Robinson et al (Lancet, Apr. 9, 1988, p. 790). The first is heat-stable, unique to HIV-1 seropositive sera, and is removed by protein-A chromatography. The second is heat-labile and ubiquitous; it is found in normal serum and is removed by heating at 60.degree. C. for 1 hour or by treatment with cobra venom anticomplementary protein. Additionally, complement component C3 deficient serum lacks the labile activity although Clq deficient serum contains the labile factor. The data suggest that the two components are antibody and the alternative pathway of complement fixation. The mechanism of action does not involve an increase in either complement-mediated cytolosis or syncytium formation. The activity has been identified in the majority of patients tested to date.
Antibody-dependent enhancement (ADE) of human immunodeficiency virus type 1 (HIV-1) infection in vitro has been described recently and shown to require participation of the alternative pathway of complement or to involve an Fc receptor-mediated mechanism. Complement-mediated ADE results in an accelerated cytopathic effect in target cells which can abrogate the protective properties of neutralizing antibodies (Robinson et al, Lancet i:693-699, 1987; Robinson et al, Lancet i:790-795, 1988). This study utilized MT-2 cell cultures to demonstrate that ADE of HIV-1 infection causes an acceleration of several parameters indicative of HIV-1 infection including: HIV-1 antigen synthesis as detected by indirect immunofluorescence, mRNA accumulation as measured by a solution hybridization protocol, reverse transcriptase release, and progeny virus production. Thus, ADE of HIV-1 infection in vitro was characterized by an authentic, accelerated rate of HIV-1 infection.
Despite the presence of antibodies which neutralize the human immunodeficiency virus (HIV-1) in vitro, such antibodies fail to protect the infected individual from the immunological sequelae that lead to the Acquired Immune Deficiency Syndrome (AIDS) and, ultimately, to death. Recently, Robinson et al reported the existence of an immunoglobulin in the serum of greater than 60% of HIV-1 seropositive individuals that, in combination with the alternative pathway of complement, can lead to a more rapid viral-induced cytopathic effect. They demonstrated that the cytopathic effect (CPE) observed in this antibody-dependent enhancement of HIV-1 infection is the direct result of a more rapid infection of the target cell line by the HIV-1. This ADE of HIV-1 infection is manifested by more rapid accumulation of HIV-1-specific RNA as well as faster expression of detectable viral proteins and accelerated production of infectious virus particles. A majority of HIV patients have been observed to have this antibody-dependent enhancement. It is believed that a majority of other viruses, such as influenza, Dengue fever and related encephalitis, that exhibit antigenic drift and/or infectious, viral escape mediated through ADE or similar mechanisms exhibit similar factors.
ADE (source being Serum designated #110) was obtained from an anonymous donor confirmed to be HIV antibody positive by immunofluorescence and Western Blot (DuPont, Wilmington, Del.). The ADE serum was heat-inactivated for one-half hour at 60.degree. C. then diluted 1:5 into growth medium prior to being assayed for ADE of HIV-1 infection. ADE accelerated cytopathic effect of HIV-1 as determined by a decrease in percent viable cells relative to control infection in the presence of Serum #110 and human complement serum. At high antibody concentration (1:20 dilution), less than 20% of MT-2 cells were viable compared to 100% cell viability in the presence of an equivalent dose of HIV-1 and 1:20 human complement serum. This enhancement of viral cytopathic effect extended to a dilution of 14,860 where MT-2 cell viability was reduced by 30%.
Phase contrast microscopy revealed that the decreased cell viability in the presence of enhancing serum and HIV-1 was accompanied by increased giant cell formation. I showed that ADE of HIV-1 infection was responsible not only for accelerated CPE but also for an accelerated progression of other parameters of HIV-1 infection including viral antigen and RNA synthesis, progeny virus production, and reverse transcriptase (RT) release.
Experiments were performed by culturing cells from the CD4.sup.+ cell line, MT-2, in the presence of HIV-1 alone, HIV-1 and human complement serum (1:20), HIV-1 and an enhancing serum plus 1:20 human complement serum, or virus-free H9 medium and enhancing serum plus 1:20 human complement serum. CD4.sup.+ cells are those cells which react (positively) to the monoclonal antibody CD4 which reacts with a specific subset of thymus-derived lymphocytes designated T4 of T helper cells. Twelve hours after challenge, cells were washed then resuspended in growth medium without added serum or virus. Cell suspensions were removed at 24, 48, 60 and 72 hours post-infection. The culture supernatants were assayed for infections virus and RT activity while the cells were assayed for HIV-1-specific antigen and HIV-1-specific RNA.
The rate of accumulation of viral antigen positive cells was increased in those cells challenged by HIV-1 in the presence of enhancing serum. For example, 48 hours post-challenge, 95% of MT-2 cells were expressing viral antigens by indirect immunofluorescence compared to only 15% and 10% of cells challenged by HIV-1 and complement of HIV-1 alone, respectively. Mock-infected cells did not express HIV-1 antigens. Therefore, ADE of HIV-1 infection was characterized by a decrease in time to detectable viral antigen synthesis.
I also observed that increases in HIV-1-specific RNA occurred sooner in cells exposed to enhancing antibody than in cells exposed to HIV-1 alone or HIV-1 and complement. Maximum quantities of HIV-1 a specific RNA for the latter two samples were never as high as the corresponding cells infected by HIV-1 in the presence of enhancing antibody. For example, at 60 hours, the cpm of hybridized probe to RNA from target cells in the ADE culture was 3631 compared to 347 for the HIV-1 alone and 374 for the HIV-1 and complement culture. It should be noted that maximum CPE in the ADE culture occurred after 60 hours so that the number of cells at the 72 hour time point was smaller than in other cultures or at other time points. The decrease in viable cell number was reflected by a decrease in HIV-1 RNA after 72 hours compared to 60 hours in the ADE culture.
If enhancing serum causes HIV-1-specific RNA and antigen to appear sooner, then one would expect increased infectious virus yields sooner, also. Infectious virus yields obtained from culture supernatants of cells infected in the presence of HIV-1 alone, HIV-1 and complement, and HIV-1 and enhancing serum with complement were evaluated.
From 24 hours after virus challenge to the endpoint at 72 hours, infectious virus yields were many times greater in cultures challenged in the presence of enhancing serum plus complement and HIV-1. At 72 hours, the infectious particles per milliliter obtained from the ADE culture were 1.3.times.10.sup.6 compared to 1.2.times.10.sup.5 for HIV-1 alone and 1.6.times.10.sup.5 from the HIV-1 plus complement culture. These data demonstrate that ADE caused in accelerated appearance of and increase in infectious virus output from the target cells.
As a final measure of virus production, culture supernatants were assayed for reverse transcriptase released from the infected cells. Again, the ADE culture exhibited the earliest release of RT activity from the infected MT-2 cells. Detectable RT activity was present as early as 24 hours post-virus challenge with a maximum achieved at the endpoint of the assay, 72 hours. At 72 hours, the RT activity in supernatants from the ADE culture was 3.3.times.10.sup.6 cpm/ml while the cells exposed to HIV-1 alone generated only 8.8.times.10.sup.5 cpm/ml. Cells exposed to HIV-1 and complement were comparable to the latter cells generating 9.6.times.10.sup.5 cpm/ml.
It should be noted that complement serum alone had the ability to activate HIV-1 virions in vitro. I have consistently observed minor decreases in time to HIV-1-specific CPE in the presence of normal human complement serum. The data support this finding since the MT-2 cells exposed to HIV-1 and complement exhibited a minor acceleration in synthesis of HIV-1 antigen, RT release, and infectious particles compared to cells exposed to HIV-1 alone. At early time points the same findings for HIV-1-specific RNA were observed although at 72 hours the cells challenged by HIV-1 alone demonstrated higher levels of RNA than the cells exposed to HIV-1 and complement serum. This was because the cells exposed to HIV-1 and complement serum had begun to exhibit lysis at that time while the cells exposed to HIV-1 alone had not.
Since virus, antibody, and complement were removed and the cells washed twelve hours after virus challenge, any effect that enhancing antibody and complement had on the virus must have occurred within the first twelve hours of virus exposure to the MT-2 cells. Even though both of the HIV-infected cultures without enhancing antibody (i.e., HIV-1 alone and HIV-1 plus complement serum) became 100% positive for HIV-1 antigen, they never came within a log of the RT activities or infectious virus released from the MT-2 cells exposed to virus in the presence of enhancing antibody.
Although not wishing to bound by any particular theory, these data can be interpreted to suggest that ADE not only increases the rate of detectable antigen expression in infected cells, but may also increase the copies of HIV-1 genome or mRNA in infected cells. This might occur either by an increase in the number of virions that infect each MT-2 cell or, alternatively, by an increase in transcriptional efficiency of the virus in the presence of enhancing serum. The latter seems unlikely since preliminary data suggests that antibodies conferring ADE of HIV-1 infection are directed toward the virus env gene products, gp41 and gp120.
In the detailed discussion of this invention (below), I demonstrate the effectiveness of dsRNA against various antigenic isolates of human retroviruses. I also show the ability of dsRNA to act in face of the above described antibody-dependent, complement- mediated enhancers of viral infection and viral escape.
Recent attempts to develop a vaccine to HIV have not been successful. All HIV-1 candidate vaccines tested in chimpanzees (animal models) to date have failed to afford protection from subsequent HIV-1 challenge. Recent reports describing antibody-dependent enhancement (ADE) of HIV-1 infection in vitro offer an intriguing possibility for vaccine failure in chimpanzees. Using an MT-2 cell assay for ADE, as described, it is demonstrated that complement-dependent enhancing antibodies are produced by chimpanzees infected by HIV-1. Further, HIV antibody-negative, fresh chimpanzee serum enhances HIV-1 infection. The data further suggest a complement-mediated ADE mechanism to explain the failure to passively immunize chimpanzees with human IgG fractions containing high HIV-1 neutralizing activity.
The majority of individuals infected with type-1 human immunodeficiency virus (HIV-1) exhibit antibodies which, in conjunction with the alternate complement pathway, accelerate HIV-1 infection of MT-2 target cells in vitro. This phenomenon, known as antibody-dependent enhancement (ADE), has been documented in other viral infectious processes (e.g., see Porterfield, Adv. Virus Res., Vol. 31, p. 335, 1986 and Halstead, Science. Vol. 239, p. 476, 1988), and has been confirmed recently for complement-dependent ADE of HIV-1 infection in peripheral blood lymphocytes as well as extended to include an apparent Fc receptor-dependent mechanism for HIV-1 ADE. The Fc receptor-dependent mechanism was demonstrated not only in sera obtained from HIV-1 infected individuals, but from HIV-1-induced antisera obtained from guinea pigs and chimpanzees. Of special concern to vaccine development is the observation that complement-dependent ADE can reduce or completely abrogate the protective effects of neutralizing antibody within the same sera as well as between non-homologous human sera. The failure of candidate HIV-1 envelope vaccines to protect chimpanzees from infection despite the presence of neutralizing antibodies (see Berman, Proc. Natl. Acad. Sci. U.S.A., Vol. 85, p. 5200, 1988) may be secondary to the concomitant induction of antibodies which promote ADE at the expense of neutralizing antibodies. Similarly, the failure to protect chimpanzees from HIV-1 infection by passive transfer of high titer human neutralizing immunoglobulin may be the result of ADE. Since the only surrogate animal available for HIV-1 infectivity studies is the chimpanzee, the demonstration of complement-dependent ADE in the chimpanzee is critical to the evaluation of candidate vaccines for the capacity to induce ADE. I have investigated by invention using HIV-1 induction of ADE in the chimpanzee and determined the comparative roles of chimpanzee versus human complement in mediation of the response by both seronegative and seropositive subjects.
Chimpanzees have been used as models for approximating human vaccine requirements. To investigate if antibody-negative chimpanzee serum could, indeed, cause an enhancement of HIV-1 infection, thereby leading to the high background CPE observed when chimpanzee serum is used as a source of complement, I performed two-fold dilutions of fresh, antibody-negative chimpanzee serum. I showed that human complement serum was capable of enhancing the infectivity of HIV-1 slightly but only to a dilution of 1:16; therefore, at a dilution of 1:20, no enhancement of infection could occur without the additional presence of antibody to HIV-1. Serum from a control chimpanzee X95 could cause an increased rate of viral-induced cytolysis to a dilution of 1:256 while serum from chimpanzee X35 could enhance infection to a dilution greater than 1:512. To determine if antibody-negative chimpanzee serum X35 was enhancing HIV-1 infection via complement-mediated mechanism, I tested the ability of cobra venom anticomplementary protein to abrogate the enhancing activity of the chimpanzee serum. I found that a one hour pre-incubation of chimpanzee serum with cobra venom factor completely inactivated serum X35, thus blocking the enhancement of HIV-1 infection in a manner analogous to ADE of HIV-1 infection by human serum. All further studies, therefore, utilized chimpanzee serum heat-inactivated one hour at 60.degree. C. plus fresh human serum as a complement source. To confirm that increased CPE was associated with increased rate of HIV-1 infection, the effect of chimpanzee complement serum on accumulation of reverse transcriptase (RT) and immunofluorescence (IFA) was determined. Chimpanzee complement serum and chimpanzee complement serum plus serum from chimpanzee X91 greatly increased both RT activity and the percentage of IFA positive cells at 42 hours post virus challenge. Serum from chimpanzee X91 alone could not increase these parameters. Chimpanzee complement serum plus heat-inactivated serum X91 showed decreased RT activity compared to chimpanzee complement serum alone because X91 was neutralizing at this concentration (1:100). Human complement serum alone slightly increased both percent IFA positive cells and RT activity while this increase was greatly enhanced by the addition of heat-inactivated serum form chimpanzee X91. Thus, enhanced HIV-1 infection by chimpanzee complement serum as well as chimpanzee antibody and human complement serum were correlated with enhancement of CPE.